System and method for radar based mapping for autonomous robotic devices

Abstract

Data from a radar sensor (502) moving through a static environment may be smoothed and used to generate range profiles by approximating peaks. A direction of arrival (DOA) can then be determined based on the range profile in order to generate a reprojection map. The reprojection map is used to provide updates to a stored map in a robot (102).

We claim: 
     
         1 . A method for mapping an environment in which an autonomous device is located, the method comprising:
 receiving raw radar data at each of three radar receivers of the autonomous device, the three radar receivers being disposed in a noncollinear arrangement, and the raw radar data corresponding to a radar signal reflected by an object in the environment;   estimating, using a processor in communication with the three radar receivers, respective peaks in the raw radar data received at each of the three radar receivers;   determining, using the processor, a Doppler speed associated with the raw radar data;   determining, using the processor, a direction of arrival (DOA) of the object based on the estimated peaks and the Doppler speed; and   generating a reprojection map based on the DOA and the estimated respective peaks.   
     
     
         2 . The method of  claim 1 , further comprising automatically selecting a direction of travel of the autonomous device based on the reprojection map. 
     
     
         3 . The method of  claim 1 , further comprising updating a pre-existing map of the environment based on the reprojection map. 
     
     
         4 . The method of  claim 1 , further comprising determining relative phase information corresponding to at least a subset of the estimated peaks, wherein determining the DOA based on the estimated peaks and the Doppler speed includes determining the DOA based in part on the relative phase information. 
     
     
         5 . The method of  claim 4 , wherein determining the relative phase information includes determining a first phase shift between the respective estimated peaks associated with a first one of the three radar receivers and a second one of the three radar receivers, and determining a second phase shift between the respective estimated peaks associated with the first one of the three radar receivers and a third one of the three radar receivers. 
     
     
         6 . The method of  claim 5 , wherein determining the DOA includes estimating a first DOA component based on the first phase shift and estimating a second DOA component based on the second phase shift. 
     
     
         7 . The method of  claim 1 , wherein at least two of the three radar receivers are spaced apart by half a wavelength or less. 
     
     
         8 . A method for mapping an environment in which a robot is located, the method comprising:
 receiving, at a plurality of receivers of the robot, a radar signal reflected by a feature of the environment, wherein the plurality of receivers includes at least three receivers arranged noncollinearly;   determining, using a processor in communication with the plurality of receivers, a Doppler speed associated with the reflected radar signal;   calculating, using the processor, a direction of arrival (DOA) of the environmental feature based on the Doppler speed and respective peaks of the reflected radar signal received at each of the plurality of receivers; and   updating a map of the environment based on the DOA and the reflected radar signal.   
     
     
         9 . The method of  claim 8 , wherein updating the map comprises generating a reprojection map based on the DOA and the peaks of the reflected radar signal, and updating a pre-existing map based on the reprojection map. 
     
     
         10 . The method of  claim 8 , further comprising controlling movement of the robot within the environment based on the updated map. 
     
     
         11 . The method of  claim 8 , wherein calculating the DOA includes calculating a first component of the DOA based on a first phase shift between the peaks associated with a first subset of the plurality of receivers, and calculating a second component of the DOA based on a second phase shift between the peaks associated with a second subset of the plurality of receivers. 
     
     
         12 . The method of  claim 11 , wherein the first subset of the plurality of receivers and the second subset of the plurality of receivers have at least one receiver in common. 
     
     
         13 . The method of  claim 8 , wherein the receivers of the plurality of receivers are spaced from one another by a distance less than or equal to half a wavelength characteristic of the reflected radar signal. 
     
     
         14 . A method of navigating an autonomous device in an environment, the method comprising:
 receiving raw radar data at each of two radar receivers of an autonomous device, the raw radar data corresponding to a radar signal reflected by an object in the environment, wherein the two radar receivers are spaced from each other by half a wavelength or less;   determining, using a processor in communication with the two radar receivers, relative phase information corresponding to peaks in the raw radar data received at the two radar receivers;   determining, using the processor, a Doppler speed associated with the raw radar data;   determining, using the processor, a direction of arrival (DOA) associated with the object based on the Doppler speed, the relative phase information, and the peaks; and   generating a reprojection map based on the DOA and the peaks.   
     
     
         15 . The method of  claim 14 , further comprising updating a pre-existing map of the environment based on the reprojection map. 
     
     
         16 . The method of  claim 14 , further comprising controlling the autonomous device to move within the environment based on the reprojection map. 
     
     
         17 . The method of  claim 14 , wherein the autonomous device further includes a third radar receiver, and the three radar receivers are disposed in a noncollinear arrangement. 
     
     
         18 . The method of  claim 17 , wherein determining the relative phase information includes:
 determining a first phase shift between the respective peaks associated with a first one of the three radar receivers and a second one of the three radar receivers, and   determining a second phase shift between the respective peaks associated with the first one of the three radar receivers and a third one of the three radar receivers.   
     
     
         19 . The method of  claim 18 , wherein determining the DOA includes estimating a horizontal DOA component based on the first phase shift and estimating a vertical DOA component based on the second phase shift. 
     
     
         20 . The method of  claim 14 , wherein the radar signal reflected by the object in the environment is a radar chirp produced by a transmitter of the autonomous device, the radar chirp having a duration in a range of 5 to 15 milliseconds.